(2) The macromolecular properties of the hyaluronidase- digested polysaccharide were studied by gel chromatography on Sephadex G-200 of the intact, as well ...
27
Biochem. J. (1970) 116, 27-34 Printed in Great Britain
Attempted Isolation of a Heparin Proteoglycan from Bovine Liver Capsule BY U. LINDAHL Institute of Medical Chemistry, University of Uppsala, Uppsala, Sweden (Received 10 July 1969)
(1) Polysaccharides were isolated from bovine liver capsule by extraction with 2M-potassium chloride followed by precipitation from 0.8M-potassium chloride with cetylpyridinium chloride. Chondroitin sulphate was eliminated by digestion with hyaluronidase. The yield of heparin was approx. 40% of that obtained after extraction of the papain-digested tissue. (2) The macromolecular properties of the hyaluronidasedigested polysaccharide were studied by gel chromatography on Sephadex G-200 of the intact, as well as of the alkali-degraded, material. The results suggested the presence of single heparin chains in addition to a dermatan sulphate proteoglycan. (3) A purified heparin preparation was analysed for amino acids and neutral sugars. Xylose, galactose and serine were found in amounts corresponding to 0.1, 0.2, and 0.4 residue/polysaccharide chain (mol.wt. 7400), respectively. It is suggested that the isolated material had been degraded by a polysaccharidase with endo-enzyme properties.
The existence of complexes between glycosaminoglyeans and proteins was first suggested by Morner (1889), who isolated a crude chondroitin 4-sulphateprotein complex ('chondromucoprotein') from cartilage. More recently proteoglycans containing chondroitin 6-sulphate (Barrett, 1968), dermatan sulphate (Toole & Lowther, 1968) and keratan sulphate (Barrett, 1968) have been isolated and partially characterized. Extraction of tissues by procedures not involving proteolysis yield proteoglyeans apparently consisting of multiple polysaccharide chains in covalent linkage to a single polypeptide (termed 'multichain proteoglycan' below). Isolation procedures including treatment of tissues with proteolytic enzymes generally increase the yield of polysaccharide; however, the proteoglyeans obtained are degraded, ultimately to single polysaccharide chains linked to residual peptides (Luscombe & Phelps, 1967). The carbohydrate-protein linkage regions of the proteoglycan molecules have attracted considerable attention. This applies particularly to the chondroitin 4-sulphate-protein linkage region, the structure of which has been established (Gregory, Laurent & Roden, 1964; Lindahl & Roden, 1966; Roden & Armand, 1966; Roden & Smith, 1966):
A conspicuous feature of this structure is the lability of the xylose-serine linkage, which is the site of cleavage in alkaline degradation of chondroitin 4-sulphate-protein to single polysaccharide chains (Anderson, Hoffman & Meyer, 1965). Heparin, although structurally closely related to other connective-tissue polysaccharides, displays certain unique characteristics, one being its intracellular location (Brimacombe & Webber, 1964). Nevertheless, the presence of residual amino acids in commercial heparin preparations strongly suggested the existence of a native heparin proteoglycan (Lindahl, Cifonelli, Lindahl & Roden, 1965). Further studies established the identity of the chondroitin 4-sulphate-protein and the heparin-protein linkage regions (Lindahl & Rod6n, 1965; Lirndahl, 1966a,b, 1967) thus indicating that any multichain heparin proteoglycan should be susceptible to alkali. Since the existence of a native heparin proteoglycan seemed to be amply verified, attempts were made to isolate this material on a preparative scale from bovine liver capsule. While this work was in progress the results of an identical project were published (Serafini-Fracassini, Durward & Floreani, 1969). However, these latter results are not confirmed by the present study.
--GlcUA->CGalNAc-->GlcUA->Gal->CGaI> MATERIALS AND METHODS
S04 polypeptide
-->Xyl*O-er\polypeptide
Materials. Bovine liver capsules were removed from the livers within 30min after the death of the animals and were stored at -20°C after a quick rinse with cold water.
U. LINDAHL
28
Heparin (prepared from pig intestinal mucosa; anticoagulant activity, 140U.S.P. units/mg) purchased from Wilson Laboratories, Chicago, Ill., U.S.A. was purified by repeated precipitation with CPC* from 1.2 M-NaCl essentially as described by Lindahl et al. (1965). A sample of dermatan sulphate from pig skin was a generous gift from Dr L. Roden, Departments of Pediatrics and Biochemistry, University of Chicago, Chicago, Ill., U.S.A. Samples of chondroitin 4-sulphate and hyaluronic acid were kindly supplied by Dr E. Wessler of this Institute. Heparinase (EC 3.2.1.19) isolated from Flavobacterium heparinum was generously given by Dr A. Linker, University of Utah College of Medicine, Salt Lake City, Utah, U.S.A. A highly purified preparation of testicular hyaluronidase (EC 3.2.1.35; 20000i.u./mg) was kindly supplied by Dr B. H6gberg, AB Leo, Halsingborg, Sweden; this material was stated to be free of proteolytic activity. Papain (EC 3.4.4.10; twice crystallized, 29.3mg of protein/ml of suspension) was purchased from Sigma Chemical Co., St Louis, Mo., U.S.A. Analytical method8. Analyses of total hexosamine were performed by a modification (Gardell, 1953) of the Elson-Morgan reaction after hydrolysis in 4M-HCI for 14h. The glucosamine/galactosamine ratio was determined by g.l.c. as described by Radhakrishnamurthy, Dalferes & Berenson (1966). Results obtained by this method were compared with those obtained with an automatic amino acid analyser. This double determination was carried out with a large number of glycosaminoglyean samples, including three of the fractions shown in Table 1. Results obtained by the two methods invariably differed by less than 4%. Uronic acid was determined as described by Bitter & Muir (1962). Sulphate was determined by the method of Antonopoulos (1962). Amino acid analyses were performed after hydrolysis in vacuo with 6M-HCI at 1100C for 24h, with a Bio-Cal model BC-200 amino acid analyser. Neutral sugars were determined as alditol acetates by g.l.c., by a modification of the procedure of Albersheim, Nevins, English & Karr (1967). Samples of heparin corresponding to l-lO,ug of xylose were hydrolysed with 2M-trifluoroacetic acid at 10000 for 4h. After the addition of 8,tg of inositol as internal standard, hydrolysates were deacidified by evaporation and passed successively through columns (0.4 cm x 1.5 cm) of Dowex 1 (X2; C1- form) and Dowex 50W (X8; HI form), both 200-400mesh. The use of these ion-exchange resins eliminated a number of unidentified peaks from the chromatograms. Sugars were reduced by treatment with 0.4mg of KBH4 in 0.2ml of M-NH3 at room temperature for 2h. After removal of excess of KBH4 as methyl borate (Albersheim et al. 1967) the alditols were acetylated by heating at 1000C for lh with 0.2 ml of a mixture (1:1, v/v) of acetic anhydride and pyridine in sealed glass tubes. The residue remaining after evaporation of the acetylation mixture was dissolved in lOI.l of chloroform. Samples (1-2,ul) were separated at 1900C in a Perkin-Elmer model 881 gas chromatograph on glass columns (2 m x 2 mm internal diameter) packed with 1.5% (w/w) ethylene glycol succinate and 1.5% (w/w) silicone oil (XF-1150) on Gas-Chrom P (100-120 mesh). Nitrogen was used as carrier gas, at a flow rate of approx. 30ml/min. Peak areas were measured with a Technicon *
Abbreviation: CPC, cetylpyridinium chloride.
1970
model AAG integrator-calculator. A hydrolysis curve for commercial heparin showed that the hydrolytic conditions chosen resulted in optimum liberation of galactose, whereas the yields of xylose were increased by approx. 10% when the time of hydrolysis was decreased to 2h. Hydrolysis for 4h of heparin samples containing known amounts of added xylose and galactose destroyed 15 and 10% of the sugars, respectively. No corrections for destruction during hydrolysis were applied. Neutral monosaccharides were subjected to paper chromatography in butanol-ethanol-water (10:3:5, by vol.) after hydrolysis of polysaccharides with 2m-trifluoroacetic acid at 1000C for 4 h. Samples were deionized as described by Lindahl & Roden (1965). Spots were detected by staining with a silver dip reagent (Smith, 1960). Electrophoresis of polysaccharides was carried out on strips of cellulose acetate in 0.1 M-barium acetate (2.7 V/cm for 6h) by the method of Wessler (1968). The electrophoretic mobility in this medium depends primarily on the structure of the polysaccharide backbone, whereas variations in sulphate content are of less importance. Before electrophoresis proteoglyeans were converted into single polysaccharide chains by treatment with alkali, as described below. Molecular weights of heparin preparations were determined by equilibrium ultracentrifugation as described by Yphantis (1964) at a concentration of 1.2mg/ml in M-NaCl. A 12mm 40 double-sector cell was used at 31410rev./min and 200C. The centrifugations were run for about 16 h. The value for partial specific volume used was 0.47 (Lasker & Stivala, 1966).
Extraction and fractionation of polyaaccharide8 A flow diagram of the preparation procedure is shown in Scheme 1. Frozen liver capsules were homogenized by passage through a bacteria press (Edebo, 1960); this procedure was repeated three times. A sample (1OOg) of homogenized material was suspended in 200ml. of water at 00C and extracted with gentle stirring for 30min. After centrifugation the extraction was repeated once more with water and three times with 200ml of 2M-KCI. Care was taken during the entire extraction procedure to keep the temperature below 40C. Samples of each extract were made 0.8M with respect to KCI and a few drops of 1% (w/v) CPC in 0.8M-KCl were added. Judging from the resulting turbidity significant amounts of polysaccharides were present only in the first KCI extract, which was therefore retained for further fractionation. The KCI extract, clarified by filtration through Celite, was mixed with an equal volume of 0.05M-tris buffer, pH9.0, and 0.5vol. of 1% CPC. The mixture, 0.8M with respect to KCI, was kept in a water bath at 300C for 30min and the resulting precipitate, collected by centrifugation, was dissolved in 50ml of 2M-KCI at 300C. (On numerous occasions, not mentioned specifically in the text, centrifugation did not effect quantitative sedimentation of CPC-precipitable material. In such instances, the solutions were clarified by filtration through Celite, and polysaccharides were recovered from the Celite by thorough extraction with 2M-KCI.) Insoluble material, probably denatured protein, was removed by filtration through a pad of Celite, which was then thoroughly
Vol. 116
ATTEMPTED ISOLATION OF HEPARIN PROTEOGLYCAN
29
Tissue homogenate Extraction with H20
Residue
Extract (discarded)
Extraction with 2 M-KCI
Residue (discarded)
Extract
Polysaccharide twice precipitated with CPC from 0 8 M-KCI
Polysaccharide (Fraction LC)
Supernatants
(discarded)
Digestion with hyaluronidase precipitation with CPC
Oligosaccharides Fraction LC-hyase Gel chromatography I
I
LC-hyase I
LC-hyrase II
Ethanol fractionation
LC-hyase-25% EtOH
LC-hyase-proteiin
~~~~~~~~~~~~~~~~~~~~~~
LC-hyase-5Wo EtOH
Ethanol fractionation
LC-hyase-18%o EtOH Scheme
1.
Flow diagram for fractionation of polysaccharides.
extracted with 2 M-KCI. The polysaccharides in the combined filtrate and extract (lOOml) were reprecipitated with CPC from 0.8M-KCI, as described above. The precipitate was suspended in 2M-NaCl in aq. 10% (v/v) ethanol at 30°C; again a portion of the material remained insoluble and was removed by passage through Celite. Polysaccharides were precipitated from the clarified solution (75ml) by the addition of 3vol. of 96% (v/v) ethanol. The precipitate (fraction LC) was collected by centrifugation, washed with ethanol and finally dissolved in 5ml of water. Dige8tion of fraction LC with hyaluronidase. Fraction LC (5ml, containing 6.1 mg of uronic acid) was mixed with 5 ml of 0.3 M-NaCl-0. 1 M-tris buffer, pH 7.0, containing 3 mg of hyaluronidase. A drop of toluene was added and the mixture was incubated at 37°C. After 5h another 3mg of enzyme was added and digestion was continued for 12 h. The slightly turbid solution was clarified by filtration through a Celite pad, which was subsequently rinsed with 10ml of 0.3M-NaCl-O.lm-tris buffer, pH7.0. The filtrate was adjusted to pH 9.0 with M-NaOH, and polysaccharide was precipitated by the addition of lOml of 1% CPC. After reprecipitation with CPC from 0.8 M-KCI the
hyaluronidase-digested fraction LC was recovered as its sodium salt (fraction LC-hyase) as described above. As in the preceding part of the preparation, CPC-precipitable material collected during recovery of fraction LC-hyase was only partially soluble in 2M-KCI or -NaCl. I8olation of poly8accharides after proteoly8i8 of ti88ue. Homogenized liver capsule (20.2g) was suspended in 50ml of 0.05M- sodium acetate buffer (pH5.5)-0.3M-NaCl0.01 m-EDTA-0.01 M-cysteine-HCl. After the addition of 2 ml of papain suspension, digestion was allowed to proceed at 65°C for 20h. The digest was passed through a Celite pad equilibrated with 0.3 M-NaCl and was then mixed with an equal volume of 1% CPC in 0.3M-NaCl. The precipitated polysaccharide was reprecipitated with CPC from 0.8M-KCI and finally converted into the sodium salt, yielding fraction LC-papain. Gel chromatography offraction LC-hyase. Fractionation of LC-hyase according to molecular size was accomplished by gel chromatography on a column (0.9 cmx 90 cm) of Sephadex G-200. Samples (1 ml) were mixed with equal volumes of 4m-KCI and elution was carried out with 2M-KCI, at a rate of approx. 4ml/h. Effluent fractions were analysed for uronic acid by the carbazole reaction.
1970
U. LINDAHL
30
Ethanol fractionation of fraction LC-hyase. Fraction LC-hyase, as the calcium salt, was fractionated with ethanol by the method of Meyer, Davidson, Linker & Hoffman (1956). The precipitated polysaccharides were converted into sodium salts and freeze-dried. A proteinrich material (LC-hyase-protein), which remained insoluble on dissolving freeze-dried LC-hyase in 5% (w/v) calcium acetate-0.5m-acetic acid, was retained for amino acid analysis. Degradation of poly8accharides. Attempts were made to effect alkaline cleavage of polysaccharide-protein bonds by mixing 0.5 ml of polysaccharide solution with an equal volume of M-NaOH. The mixture was kept at 4°C for 20h and was then neutralized. Digestion with papain was performed as described above, after freeze-drying of polysaccharide solutions. Digestion of polysaccharides with heparinase was carried out as described by Linker & Hovingh (1965). Digestion mixtures were passed through columns of Dowex 50W (X2; HI form), evaporated to dryness and spotted on Whatman no. 1 paper. Papers were developed with ethyl acetate-acetic acid-water (3:1:1, by vol.) and stained with a silver dip reagent (Smith, 1960).
completely eliminated by digestion with hyaluronidase, as demonstrated by electrophoresis of fraction LC-hyase (Fig. 1). The results in Table 1 show that digestion with papain drastically increases the yield of polysaccharides from capsule tissue. However, although less than 20% of the total polysaccharide content of
C-4-S * LC * Hep LC-hyase *-
*-
0 *
HA+DS C-4-S
Fig. 1. Electrophoresis of polysaccharides in barium acetate. C-4-S, chondroitin 4-sulphate; Hep, heparin; HA, hyaluronic acid, DS, dermatan sulphate.
RESULTS
Isolatw>on and identification of polysaccharides Yields and analytical results for some of the polysaccharide fractions are shown in Tables 1 and 2. Since the various fractions had been repeatedly precipitated with CPC from 0.8m-potassium chloride the presence of hyaluronic acid could be excluded (Scott, 1960). Thus, the glucosamine contents of fractions LC-papain, LC and LC-hyase must represent heparin or heparan sulphate; the uronic acid/hexosamine molar ratios indicate the absence of keratan sulphate or glucosamine-containing glycoproteins. Electrophoresis in barium acetate of fraction LC revealed the presence of three components that migrated like heparin, dermatan sulphate and chondroitin sulphate (Fig. 2). The last component was
0 .e
Table 1. Analysis of polysaccharide fractionm The glucosamine/galactosamine ratios of fraction LC and its subfraction LC-hyase refer to different preparations of liver-capsule polysaccharides. The remaining subfractions are all derived from the LC preparation shown in the table. Yield (mg of uronic Ratio uronic Gluc bosamine total acid/ acid/lOOg (exoofsamine) Fraction of tissue) hexosamine hexoi 35.9 1.34 34 LC-papain 6.1 72 LC 1.30 1.22 5.6 75 LC-hyase 47 LC-hyase I 95 LC-hyase II
20
30
40
50
60
t Elution volume (ml) Fig. 2. Chromatography on Sephadex G-200 of: (a) fraction LC-hyase (-) and commercial heparin (x); (b) alkalitreated LC-hyase I; (c) alkali-treated LC-hyase II. Effluent fractions were analysed for uronic acid by the carbazole reaction.
VO
ATTEMPTED ISOLATION OF HEPARIN PROTEOGLYCAN 31 fraction LC-papain was recovered in fraction LC [not Table 2. Analysis offraction LC-hyase-50% EtOH Vol. 116
considering the somewhat differing colour yields of iduronic acid and glucuronic acid in the carbazole reaction (Bitter & Muir, 1962)], the latter fraction accounted for about 40% of the glucosamine and thus, presumably, the heparin content of the papaindigested fraction. Apparently, heparin is extracted with greater ease than the galactosamine-containing polysaccharides. The yield of fraction LC should be evaluated in the light of the difficulty of removing large amounts of denatured protein without concomitant loss of polysaccharide. Fraction LC-hyase was further fractionated by gel chromatography on Sephadex G-200, yielding LC-hyase I and II (Fig. 2a). Electrophoresis in barium acetate of LC-hyase I, which emerged at the void volume (V0) of the column, showed the presence of heparin and dermatan sulphate. Distinct spots appeared only when the material had been subjected to alkaline conditions (see the Materials and Methods section) before electrophoresis. Fraction LC-hyase II was retarded relative to a sample of commercial heparin on gel chromatography (Fig. 2a). Electrophoresis showed one major spot that had migrated like heparin, along with a trace of dermatan sulphate. The presence in LC-hyase I of approximately equal amounts of heparin and dermatan sulphate, as suggested by electrophoresis, was confirmed by hexosamine analysis (Table 1), which showed almost equimolar amounts of glucosamine and galactosamine. The glucosamine content of LC-hyase II indicated heparin as the major polysaccharide constituent. An attempt to achieve complete separation of the heparin and dermatan sulphate of Fraction LC-hyase by ethanol fractionation was unsuccessful. Material precipitated by 25% (v/v) ethanol was mainly heparin, since glucosamine constituted 77% of the total hexosamine content. On reprecipitation of this material (LC-hyase-25% EtOH; Scheme 1) from 18% (v/v) ethanol (LC-hyase-18% EtOH) the corresponding value decreased to 69%. Dermatan sulphate free of heparin was thus not obtained. On the other hand, dermatan sulphate was quantitatively eliminated from fraction LC-hyase by precipitation from 25% ethanol, since the remaining polysaccharide material (LC-hyase-50% EtOH) was practically devoid of galactosamine (Table 2) and showed only heparin on electrophoresis in barium acetate. In addition to these observations fraction LC-hyase50% EtOH, which accounted for about 40% of the glucosamine content of fraction LC-hyase, was identified as heparin by the following criteria: (a) resistance towards hyaluronidase; (b) high sulphate content (Table 2); (c) susceptibility to heparinase. Paper chromatography of a heparinase digest showed a pattern of low-molecular weight products identical with that obtained with a digest of commercial heparin.
Uronic acid (% of material*) Hexosamine (% of material*) Glucosamine (% of total hexosamine) Sulphate/hexosamine molar ratio
Nw *
25.8 20.2 98 2.2 7400
Not corrected for moisture
Macronmolecular properties of polysaccharides The retarded position of peak II in the gel chromatogram of fraction LC-hyase (Fig. 2a) indicated that the corresponding polysaccharide occurred as single chains and not as a multichain proteoglycan. Accordingly, the elution position of alkali-treated peak II was the same as that ofthe untreated material (Fig. 2c). Alkali treatment of peak I, on the other hand, resulted in a marked degradation of the material (Fig. 2b), indicating the breakdown of a multichain proteoglycan to single polysaccharide chains. However, since part of the alkali-degraded material still emerged at the void volume of the column, the possibility remained that only one of the two polysaccharides, heparin and dermatan sulphate, present in LC-hyase I had been affected by the alkaline conditions. To clarify this point the alkali-treated LC-hyase I was separated into a higher- and a lowermolecular-weight portion, LC-hyase Ia and Ib, respectively (Fig. 2b). Fraction LC-hyase Ia consisted of both heparin and dermatan sulphate, as indicated by electrophoresis, whereas the polysaccharide of fraction LC-hyase lb wasalmostexclusively dermatan sulphate. Considering the elution behaviour of the single heparin chains (Fig. 2a) the predicted elution position of heparin liberated from a multichain proteoglycan would be that of fraction LC-hyase lb. The absence of heparin in this fraction therefore strongly indicates that fraction LC-hyase I does not contain any significant amounts of alkali-labile multichain heparin proteoglyean. It is concluded that fraction LC-hyase I consists of a dermatan sulphate multichain proteoglycan along with single heparin chains or heparin-peptides of relatively high molecular weight. The presence of single chains of dermatan sulphate as well cannot be excluded.
Amino acids and neutral sugars infraction LC-hyase50% EtOH The low molecular weight of Fraction LC-hyase50% EtOH indicates that this material is essentially the equivalent of fraction LC-hyase II. However, since fraction LC-hyase-50% EtOH seemed to contain heparin of somewhat higher purity than fraction LC-hyase II (Tables 1 and 2) the former material was used for amino acid and neutral sugar
1970
U. LINDAHL
32 0.7 r C)
0.7 r
(a)
(b)
0.6 F
0.6 i 0
'C
C
0.5 *
0.5 F
C -4-
;-C
0.4
0.4 F C.)
0.3 F
C1)
0.2 F
C~ 'c
9 Q._
I
S 'CD 0 Cd C.)
¢e
I
O.
(d)
(c)
k
0.3F
0.2k 0. F 0
0
